Statistical Survey on alchemy mechanics

[gnox24]

Here are simulated test runs with my models of the cooker/alembic:

The numbers show that there is an overall purity gain for lower purity items. Higher purity items suffer from loss of overall purity.

Interleaved application of cooker/alembic should bring better results. This is due to the fact that the alembic removes less purity on lower purity levels - The less and earlier it's used, the better the result.

I'm hope for better results if the Test Tubes are mixed in, but I don't have much hope. Using them is a gamble, which can only be contered by hand selecting the better results.


low purity start:

Code: Select all

                                                        Purity

Start:      30.0000   25.0000   25.0000   20.0000       0.6667

Cooker:     30.0000   26.0000   24.4444   19.5500       0.7431
            30.0000   27.0000   23.8888   19.0999       0.8598
            30.0000   28.0000   23.3331   18.6498       1.0167
            30.0000   29.0000   22.7773   18.1996       1.2138
            30.0000   30.0000   22.2215   17.7494       1.4511

Alembic:    29.6875   29.6875   22.3951   18.2026       1.2743
            29.3945   29.3945   22.5579   18.6274       1.1189
            29.1199   29.1199   22.7106   19.0257       0.9824
            28.8624   28.8624   22.8537   19.3991       0.8625
            28.6210   28.6210   22.9878   19.7491       0.7572


Mediocre purity start:

Code: Select all

                                                        Purtiy

Start:      60.0000   20.0000   15.0000    5.0000       23.3333

Cooker:     60.0000   21.0000   14.2500    4.7403       23.5538
            60.0000   22.0000   13.4996    4.4804       23.8175
            60.0000   23.0000   12.7488    4.2203       24.1246
            60.0000   24.0000   11.9975    3.9601       24.4750
            60.0000   25.0000   11.2457    3.6997       24.8687

Alembic:    57.8125   25.0000   12.1053    5.0309       21.8551
            55.7617   25.0000   12.9112    6.2790       19.2066
            53.8391   25.0000   13.6668    7.4491       16.8789
            52.0367   25.0000   14.3751    8.5460       14.8332
            50.3469   25.0000   15.0392    9.5744       13.0354


[Snowpig]

Apart from these values not looking very promising you have made an error in the last set of data:

Code: Select all

Alembic:    28.8624   28.4129   23.1239   19.5814        0.7797
Cooker:     28.8624   29.4129   22.5824   19.1170        0.9811
Alembic:    28.6210   29.1371   22.7335   19.4847        0.8613

the cooker would set the values not like in second line, but more like:

Code: Select all

28.8624   28.8624 ...

[gnox24]

Ah thanks. My code doesn't check for limits at the moment. I'll remove that from the post.

[gnox24]

This survey is actually over, because yesterday evening I ctrl-clicked (instead of shift-clicked) my whole alchemy equipment and logged off. Now all is gone and i dont want to invest more silver.

But I have some final results:

Pressure Cooker's initial instability increment appears to be ~45% for 'a bit' and ~55% for 'slightly'. Afterwards there is a ~40% probability that it increases about 2 levels. This is the only thing of the four which is actually usefull. While raising purity levels of pots, compost bins, crop, etc. I use it a lot. For every step i put the source items as well as the itermediate products into my cookers to gain a little bit of purity. I even cook cookers to gain a little bit of speed.

Alembic has actually only a 20% - 25% probability to drop 2 instability levels, which is quite low and more or less completely negates the effect of the other three types. It's far to less effective to be usefull.

The random effect of the Test Tubes appears to be normal distributed with a mean of 0% and a standard deviation around ~1.1%. Together with the long tick time (around 4 hours) and the high instability increase (3, sometimes 2 levels) it's pretty useless as well. I can't think of an usefull application.

I didn't do test runs for the Conjunctive Retort because it's around three times as expensive as the others. I guess it behaves like the pressure cooker does, but i'm not sure.


As a general conclusion i would say: Pressure cook, pressure cook, pressure cook and skip the others.

[MagicManICT]

The random effect of the Test Tubes appears to be normal distributed with a mean of 0% and a standard deviation around ~1.1%. Together with the long tick time (around 4 hours) and the high instability increase (3, sometimes 2 levels) it's pretty useless as well. I can't think of an usefull application.

My only possible guess here would be if you have a relatively high area in one element to try and get it to mutate into one of the others without a huge decrease in purity or loss of a large amount of time/materials. Right now, I'll agree that it's mostly worthless.

[gnox24]

My only possible guess here would be if you have a relatively high area in one element to try and get it to mutate into one of the others without a huge decrease in purity or loss of a large amount of time/materials. Right now, I'll agree that it's mostly worthless.

Yes that could be some point to start with, but the problem is: After you have done this (hopefully in one tick) you have to put the item into the alembic, since it's so instable that it will likely be destroyed in the next tick of whatever you use it in.

Maybe it's usable for raising one element while gardening. Plant like 3 plants and put the resulting 6(9 with horti) into testubes. Now there is a 50% probabilty that your main element has risen. These can be replanted again or be fed to worms - increasing the overall value in the next run. It's a pretty slow process, but without the alchemy equipment it's even slower.

[Potjeh]

Yeah, test tubes seem like the only way to push the highest element.

[Spazzmaticus]

For pots the general idea for me has been to just mass produce. Get good lime, granite and clay. Put it all in the test tubes and hope for the best. Statistically speaking you should end up with at least a few that are increased slightly in the element you want. Now make an unburnt pot and cram THAT into the tubes. Hope for a good increase in the right element.

The unburnt pot you make will always be stable even if the individual ingredients are not. With only 1 tick in the test tubes you'll never run the risk of losing equipment, only the item being randomized.

This approach is only viable if you have a liberal amount of raw materials and can afford to lose most of them. Not a terrible way of doing things if you're aiming to get a single element high.